Microscope setup for resonance fluorescence experiments on a single InGaAs quantum dot. Two-level systems implemented in different materials can be studied both at room temperature and at low temperature. Here, a setup to probe semiconductor quantum dots is shown. The samples with a hemispherical solid immersion lens (SIL) and the objective lens are located inside a bath cryostat, the rest of the microscope remains at room temperature. Optical access is provided by a sealed laser window. The microscope design: three modules, the lower horizontal microscope arm, the vertical arm, and the upper horizontal arm, are fixed to a central cage containing two polarizing beam splitters (PBSs). The excitation laser is injected via the lower horizontal arm; the vertical arm is used for detection; and the upper horizontal arm for imaging the sample surface. Optical fibres connect the microscope to lasers and detectors mounted on an adjacent optical table. Laser suppression is implemented by means of orthogonal excitation/collection polarization states: the linear polarizer sets the laser polarization to s, matching the lower PBS; the quarter-wave plate controls the state of polarization; and the PBSs reject the s-polarized back-reflected laser light. Solid lines indicate s-polarization, dashed lines indicate p-polarization.
Laser suppression. Reflected laser light is monitored at two different angles θ of the quarter-wave plate. At θRF + π/4 (a) the laser is optimally transmitted, at θ = θRF (b) it is optimally filtered. The count rate decreases from to (corrected for dark counts), corresponding to a laser suppression exceeding 8 orders of magnitude. A silicon avalanche photodiode in photon counting mode with a dark count rate of 14 Hz (c) is used to detect the laser light reflected from a quantum dot sample (GaAs plus thin metal layer, reflectivity ∼50%). The mean laser count rate ( ) is less than the mean dark count rate ( ). Integration time per point 0.1 s.
Sensitivity of the laser suppression to the quarter-wave plate angle. The laser light reflected at the quantum dot sample (GaAs plus thin metal layer) is recorded by a silicon avalanche photodiode in photon counting mode as the quarter-wave plate angle is varied, and the corresponding optical density is calculated. At the angle of optimum laser rejection (OD > 8), a change in angle of only 2.5 mdeg causes the OD to decrease by one order of magnitude.
Long-term behaviour. The microscope is aligned to reject the laser reflected at the quantum dot sample (GaAs plus thin metal layer), and the residual counts are recorded by a single photon detector. The optical density (OD), defined as OD = −log (1/T) with transmission T, is plotted as a function of time. The microscope is stable over many hours with an .
Resonance fluorescence on a single InGaAs quantum dot with different optical Rabi couplings. Resonance fluorescence spectra are recorded with a single photon detector at constant laser frequency. Detuning is achieved by sweeping the gate voltage with respect to the laser frequency. (a) Below quantum dot saturation, at an excitation power corresponding to a Rabi energy Ω of 0.7 μeV, a signal-to-background ratio of 39 000:1 is achieved. (b) At high pump power, where power broadening dominates the optical linewidth, a signal-to-background ratio >103:1 is realized. Solid red lines show Lorentzian fits to the data (black points), blue dashed lines indicate the background.
g (2) measurement of the resonance fluorescence from the neutral exciton X0 in a single InGaAs quantum dot. A clear dip at zero time delay demonstrates photon anti-bunching. The red curve shows the convolution of the two-level atom result, 28 with λ = (Ω2 − 1/4τ2)1/2, Rabi frequency Ω, and radiative lifetime τ, with the response of the detectors (Gaussian with FWHM 0.67 ns) and provides a very good description of the data (black points). The blue curve shows the two-level atom response only. A lifetime of and a Rabi frequency were determined by fitting the data to the convolution. The measurement time was 9 h with a single channel count rate of 250 kHz.
Resonance fluorescence spectra of a single InGaAs quantum dot in a magnetic field. The laser suppression at high magnetic field is as good as that achieved at zero magnetic field. At the lineshape is a top hat and there is a hysteresis between forward and backward scanning directions. This effect is referred to as dragging. 30,31
Article metrics loading...
Full text loading...